A total of 34 pairs of frozen PTC tissues and paracancerous tissues were obtained from patients (mean age, 45 years; age range, 20–65 years; 7 male patients and 27 female patients) who underwent surgical resection at The First Affiliated Hospital of Anhui Medical University (Hefei, China) between February 2021 and June 2022. No patients received radiotherapy or chemotherapy before surgery, and there was no history of neck radiation exposure during adolescence or childhood.

The thyroid cancer cell line BCPAP, the PTC cell line TPC-1 and the normal human thyroid cell line Nthy-ori 3–1 were purchased from iCell Bioscience Inc., and were cultured in RPMI-1640 medium (Beijing Solarbio Science & Technology Co., Ltd.) supplemented with 10% fetal bovine serum (FBS; Zhejiang Tianhang Biotechnology Co., Ltd.). Another PTC cell line (IHH-4) was obtained from Nanjing CoBioer Biotechnology Co., Ltd. and was maintained in DMEM (Wuhan Servicebio Technology Co., Ltd.) supplemented with 10% FBS, penicillin (100 U/ml) and streptomycin (100 mg/ml). All cells were incubated at 37°C in an incubator containing 5% CO₂. Profiling and authentication of BCPAP, TPC-1, IHH-4 and Nthy-ori 3–1 cell lines was performed using the short tandem repeat identification criteria developed by the International Cell Line Authentication Committee (18).

The EGFP-expressing lentivirus vector pLKO.1-EGFP-puro was purchased from Hunan Fenghui Biotechnology Co., Ltd. (cat. no. FH1717), whereas the GFP-expressing pLC5-ciR lentivirus vector was obtained from Guangzhou Geneseed Biotech Co., Ltd. The short hairpin RNA (shRNA) against hsa_circ_0004846 (sh-circRNA; sense, 5′-CAAGAATCATTAACCAATGGC-3′ and antisense, 5′-GCCATTGGTTAATGATTCTTG-3′) and the corresponding negative control shRNA (sh-NC; sense, 5′-TTCTCCGAACGTGTCACGT-3′ and antisense, 5′-ACGTGACACGTTCGGAGAA-3′) were obtained from General Biology (Anhui) Co., Ltd., and were sub-cloned into the pLKO.1-EGFP-puro lentivirus vector. The full-length hsa_circ_0004846 sequence was sub-cloned into the pLC5-ciR lentivirus vector (Lv-circRNA). The empty pLC5-ciR was used as a negative control (Lv-Vector).

To obtain lentiviral particles, the 2nd lentiviral generation system was used. Briefly, 293T cells (Cellverse Co., Ltd.) were transfected with 14 µg lentiviral plasmids, 10.5 µg packaging plasmids and 3.5 µg envelope plasmids using Lipofectamine™ 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) for 6 h at 37°C. After replacing the medium with fresh medium, the lentiviral particles were collected after continued incubation for 72 h at 37°C. For stable transduction, the aforementioned lentivirus particles (Lv-sh-NC, Lv-sh-circRNA, Lv-Vector and Lv-circRNA) were used to infect TPC-1 (MOI=10) and IHH-4 (MOI=20) cells. After 24 h, TPC-1 cells were selected with 2 µg/ml puromycin and IHH-4 cells were selected with 3 µg/ml puromycin for 7 days. The maintenance concentration of puromycin used for TPC-1 and IHH-4 cells was 1 and 1.5 µg/ml, respectively. Subsequently, a BX53 fluorescence microscope (Olympus Corporation) was used to confirm the transduction efficiency. Once the stably transduced cells were constructed, the cells underwent subsequent experiments according to the experimental protocol.

miR-142-3p-mimic, NC-mimic, PELI1 small interfering (si)RNA (si-PEL1) and the siRNA negative control (si-NC) were purchased from General Biology (Anhui) Co., Ltd. The sequences were as follows: miR-142-3p-mimic sense, 5′-UGUAGUGUUUCCUACUUUAUGGA-3′, antisense, 5′-UCCAUAAAGUAGGAAACACUACA-3′; NC-mimic sense, 5′-UUCUCCGAACGUGUCACGUTT-3′, antisense, 5′-ACGUGACACGUUCGGAGAATT-3′; si-PELI1 sense, 5′-CAGUCAGUACAAAGCACUAUATT-3′, antisense, UAUAGUGCUUUGUACUGACUGTT; and si-NC sense, 5′-UUCUCCGAACGUGUCACGUTT-3′, antisense, 5′-ACGUGACACGUUCGGAGAATT-3′. TPC-1 cells were transfected with the indicated mimics (75 pmol) and siRNAs (75 pmol) using Lipofectamine 3000 for 15 min at 37°C. according to the manufacturer's protocol. A total of 48 h after transfection, the cells were collected for subsequent experiments.

Total RNA was extracted from cells and tissues using the TRIpure Reagent (BioTeke Corporation), and was then reverse transcribed into the corresponding cDNA using the BeyoRT II M-MLV reverse transcriptase (Beyotime Institute of Biotechnology) or the miRNA First Strand cDNA Synthesis (Tailing Reaction) Kit (cat. no. B532451; Sangon Biotech Co., Ltd.) according to the manufacturers' protocols. qPCR was performed on the Exicycler™ 96 system (Bioneer Corporation) using SYBR Green (Beijing Solarbio Science & Technology Co., Ltd.) and 2X Taq PCR MasterMix kit (Beijing Solarbio Science & Technology Co., Ltd.). The cycling conditions comprised an initial denaturation step for 5 min at 95°C, followed by 40 cycles at 95°C for 10 sec (denaturation), annealing at 60°C for 10 sec and elongation at 72°C for 15 sec. The expression levels of the target mRNA and circRNA were normalized to the housekeeping gene β-actin, whereas miR-142-3p expression was normalized to U6. The relative quantification of target genes in frozen clinical PTC/paracancerous tissues and cultured cells was calculated using the 2^−ΔCq and 2^−ΔΔCq methods, respectively (19). The primer sequences were as follows: hsa_circ_0004846 forward, 5′-CGGGATTCTGGGATTTG-3′; reverse, 5′-TTGGGCAGCAGTTTCAT-3′; PELI1 forward, 5′-CTACCGTGAAGCATTTA-3′; reverse, 5′-ACAGAGGAACATAGGGA-3′; β-actin forward, 5′-GGCACCCAGCACAATGAA-3′; reverse, 5′-TAGAAGCATTTGCGGTGG-3′; and miR-142-3p forward, 5′-TGTAGTGTTTCCTACTTTATGGA-3′. The U6 primers and the miR-142-3p reverse primer were contained in the corresponding miRNA First Strand cDNA Synthesis (Tailing Reaction) kit for RT.

Total RNA was extracted from TPC-1 cells using the TRIpure Reagent, and was then reverse transcribed into the corresponding cDNA using the BeyoRT II M-MLV reverse transcriptase according to manufacturer's protocol. Genomic DNA (gDNA) was extracted from TPC-1 cells by using a gDNA extraction kit (BioTeke Corporation). To validate the formation of hsa_circ_0004846, the divergent primer (DP) sequences that amplify only circular transcripts and the convergent primer (CP) sequences that detect linear RNA molecules were designed to amplify linear mRNA (SAMD4A; Fig. 1A) and hsa_circ_0004846 using cDNA and gDNA as templates. β-actin was employed as the internal control. The sequences of the primers were as follows: hsa_circ_0004846 CP (forward, 5′-GGGATTCTGGGATTTGC-3′; reverse, 5′-CGTATTGATTGTGGTGGG-3′); hsa_circ_0004846 DP (forward, 5′-CGGGATTCTGGGATTTG-3′; reverse, 5′-TTGGGCAGCAGTTTCAT-3′); β-actin CP (forward, 5′-GGCACCCAGCACAATGAA-3′; reverse, 5′-TAGAAGCATTTGCGGTGG-3′); and the DP for β-actin (forward, 5′-GCCTCGCTGTCCACCTT-3′; reverse, 5′-TCTGACCCATGCCCACC-3′), which was used as a negative control. Semi-quantitative PCR was carried out on the Exicycler™ 96 machine (Bioneer Corporation) in a 20-µl reaction system containing 1 µl cDNA or gDNA, 10 µl 2X Taq PCR MasterMix and 1 µl forward/reverse primer for divergent primers (DPs) and convergent primers (CPs), respectively. The thermocycling condition were as follows: 2 min at 94°C for pre-denaturation; 40 cycles at 94°C for 10 sec for denaturation, 54°C for 20 sec for annealing and 72°C for 30 sec for extension. β-actin was used as a loading control. Subsequently, 1.5% agarose gel (Biowest) was used for electrophoresis of PCR products, and Gold View dye (Beijing Solarbio Science & Technology Co., Ltd.) was used for visualization.

Total protein extracts were isolated from cells using the RIPA lysis buffer (Beijing Solarbio Science & Technology Co., Ltd.) and were quantified using the BCA Protein Assay Kit (Beijing Solarbio Science & Technology Co., Ltd.). Western blot analysis was performed as previously described (12). The primary antibodies used were as follows: Anti-phosphorylated (p)-AKT (cat. no. AP0140; dilution, 1:500), anti-AKT (cat. no. A17909; dilution, 1:1,000), anti-p-glycogen synthase kinase-3β (GSK-3β; cat. no. AP0039; dilution, 1:500) and anti-GSK-3β (cat. no. A2081; dilution, 1:1,000) (all from ABclonal Biotech Co., Ltd). The following secondary antibodies were purchased from Beijing Solarbio Science & Technology Co., Ltd.: Goat anti-rabbit IgG/HRP (cat. no. SE134; dilution, 1:3,000) and goat anti-mouse IgG/HRP (cat. no. SE131; dilution, 1:3,000). GAPDH (cat. no. 60004-1-Ig; dilution, 1:20,000; Wuhan Sanying Biotechnology) was used as the internal control.

Cell proliferation and viability were assessed using a CCK-8 Assay Kit (Nanjing KeyGen Biotech Co., Ltd.), according to the manufacturer's instructions. Briefly, after being transfected for 48 h, 5×10³ cells/well were seeded into 96-well plates and incubated for 24, 48 or 72 h. Subsequently, each well was supplemented with 10 µl CCK-8 solution and incubated for an additional 2 h. Finally, the absorbance in each well was measured at a wavelength of 450 nm using an 800TS Microplate Reader (Biotek; Agilent Technologies, Inc.).

The colony formation assay was performed as previously described (20). Briefly, TPC-1 and IHH-4 cells transduced with the corresponding lentiviral particles were seeded into culture dishes (400 cells/dish) and maintained in an incubator for 14 days. After washing with PBS, cells were fixed with 4% paraformaldehyde for 1 min at room temperature and then stained with Wright-Giemsa solution (Nanjing Jiancheng Bioengineering Institute) for 5 min at room temperature. The visible colonies containing ≥50 cells were manually counted under an IX53 light microscope (Olympus Corporation). The colony formation rate was calculated using the following formula: Colony formation rate (%)=(number of colonies/400) ×100.

The EdU Imaging Detection Kit (Nanjing KeyGen Biotech Co., Ltd.) was utilized to evaluate cell proliferation, according to the manufacturer's instructions. Briefly, transduced TPC-1 and IHH-4 cells were treated with 10 µM EdU staining solution at 37°C for 2 h, followed by fixing with 4% paraformaldehyde for 15 min at room temperature and staining with the Click-iT reaction solution for 30 min at room temperature. The cell nuclei were stained with DAPI for 5 min at room temperature. Finally, images of the cells were captured under an IX53 light microscope (Olympus Corporation).

To detect the protein expression levels of vimentin in PTC cells, IF staining was conducted. Specifically, TPC-1 and IHH-4 cells were fixed with 4% paraformaldehyde for 15 min at room temperature. After washing with PBS, cells were permeabilized with 0.1% Triton X-100 (Beyotime Institute of Biotechnology) for 30 min at room temperature. Subsequently, the slides were blocked with 1% bovine serum albumin (Sangon Biotech Co., Ltd.) for 15 min at room temperature and incubated with a primary antibody against vimentin (cat. no. A19607; dilution, 1:100; ABclonal Biotech Co., Ltd) overnight at 4°C, followed by incubation with Goat anti-Rabbit IgG (Heavy chain) Superclonal™ Secondary Antibody, Alexa Fluor™ 555 (cat. no. A27039; dilution, 1:200; Invitrogen; Thermo Fisher Scientific, Inc.) for 1 h at room temperature. Subsequently, the cell nuclei were stained with DAPI for 5 min at room temperature and images were captured at magnification of ×400 under a BX53 fluorescence microscope (Olympus Corporation).

For the cell migration assay, 300 µl cell suspension containing 5×10³ PTC cells was added to the upper Transwell chambers (24 wells; pore size, 8.0 µm). For the cell invasion assay, 5×10⁴ PTC cells in 300 µl were added to the upper chambers, which were precoated with Matrigel for 2 h at 37°C. The lower chamber was supplemented with 700 µl culture medium containing 10% FBS. Subsequently, the chambers were incubated in a 37°C incubator with 5% CO₂. Following incubation for 24 h, the Transwell chambers were washed with PBS, fixed in 4% paraformaldehyde for 20 min at room temperature and stained with 0.5% crystal violet solution (Amresco, LLC) for 5 min at room temperature. Images of the migratory and invasive cells were captured under a light microscope (IX53; Olympus Corporation).

For the dual-luciferase reporter assay, the wild-type (WT) and mutant (Mut) sequences of the 3′untranslated region (3′UTR) of PELI1 (WT-PELI1 3′UTR or Mut-PELI1 3′UTR), and WT-hsa_circ_0004846 and Mut-hsa_circ_0004846 fragments were sub-cloned into the pmirGLO vector [General Biology (Anhui) Co., Ltd.]. 293T cells were co-transfected with pmirGLO vectors (pmirGLO-WT-PELI1-3′UTR, pmirGLO-Mut-PELI1-3′UTR, pmirGLO-WT-hsa_circ_0004846 or pmirGLO-hsa_circ_0004846) and miRNA mimics (miR-142-3p-mimic or NC-mimic) using Lipofectamine 3000. After 48 h, firefly and Renilla luciferase activities were detected using a dual-luciferase reporter gene assay kit (Nanjing KeyGen Biotech Co., Ltd.). The relative luciferase activity was normalized to Renilla luciferase activity.

The circRNA Interactome database (https://circinteractome.nia.nih.gov/) was used to predict the downstream targets for hsa_circ_0004846, while an online target prediction (http://www.mirdb.org/) was performed to discover the downstream targets for miR-142-3p.

Data are presented as the mean ± SD. The in vitro experiments were repeated at last three times. Statistical analysis was performed using GraphPad Prism 9.0 software (Dotmatics). The paired Student's t-test was used to assess the differences between the paired PTC tumor and paracancerous tissues. Unless otherwise specified, the differences between two groups were compared by unpaired Student's t-test (two-tailed), while those among multiple groups were compared using one-way ANOVA followed by Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

hsa_circ_0004846 (length, 519 nucleotides; chr14: 55168779-55169298) is derived from back-splicing of SAMD4A mRNA (Fig. 1A). Firstly, the specificity of the primers for hsa_circ_0004846 were verified by RT-qPCR. As shown in Fig. 1B, convergent primers amplified both cDNA and gDNA, whereas divergent primers detected hsa_circ_0004846 in cDNA but not in gDNA. Subsequently, the analysis revealed that hsa_circ_0004846 was more highly expressed in PTC tissues compared with that in paracancerous tissues (Fig. 1C). Consistently, hsa_circ_0004846 was also upregulated in the thyroid cancer cell line BCPAP and the two PTC cell lines, TPC-1 and IHH-4, compared with that in the Nthy-ori 3–1 normal human thyroid cell line (Fig. 1D). These findings encouraged further investigation of the role of hsa_circ_0004846 in PTC. Therefore, hsa_circ_0004846-overexpressing (Lv-circRNA) and -depleted (Lv-sh-circRNA) TPC-1 and IHH-4 cell lines were established via lentiviral infection. The transduction efficiency was confirmed by fluorescence detection and RT-qPCR analysis (Fig. 1E and F).

As shown in Fig. 2A and B, TPC-1 and IHH-4 cell proliferation was significantly enhanced following hsa_circ_0004846 overexpression; however, proliferation was markedly reduced in hsa_circ_0004846-depleted PTC cells. Similar results were observed using the colony formation assay (Fig. 2C and D). EdU staining further verified that hsa_circ_0004846 overexpression enhanced PTC cell proliferation, whereas hsa_circ_0004846 knockdown mitigated this process (Fig. 2E and F). Furthermore, Transwell assays showed that hsa_circ_0004846 knockdown significantly attenuated PTC cell migration and invasion (Fig. 3A and B); however, TPC-1 and IHH-4 cells transduced with the hsa_circ_0004846 overexpression lentivirus displayed stronger migratory and invasive capabilities (Fig. 3C and D). Consistently, as shown in Fig. 3E, vimentin loss was observed in hsa_circ_0004846-silenced cells, whereas hsa_circ_0004846-overexpressing cells displayed the opposite effect.

Given that the PI3K/AKT pathway serves a notable role in the pathogenesis of PTC (21), the current study further explored whether hsa_circ_0004846-induced PTC cell proliferation was mediated by activation of the PI3K/AKT pathway. As shown in Fig. 4A and B, hsa_circ_0004846 overexpression in TPC-1 and IHH-4 cells upregulated p-AKT and p-GSK-3β. By contrast, hsa_circ_0004846 knockdown notably suppressed the expression levels of p-AKT and p-GSK-3β. However, hsa_circ_0004846 overexpression or silencing had no effect on the protein expression levels of total AKT and GSK-3β.

Bioinformatics analysis was performed using the circRNA Interactome database. The analysis predicted that hsa_circ_0004846 could bind to miR-142-3p (Fig. 5A-a). Dual-luciferase reporter assays verified that the luciferase activity was significantly decreased in cells co-transfected with miR-142-3p-mimic and WT-circRNA compared with that in the NC-mimic + WT-circRNA group (Fig. 5A-b). Furthermore, RT-qPCR analysis indicated that hsa_circ_0004846 knockdown could upregulate miR-142-3p and downregulate PELI1 in TPC-1 cells (Fig. 5B and C). By contrast, downregulated miR-142-3p and upregulated PELI1 expression were observed in hsa_circ_0004846-overexpressing cells. In addition, bioinformatics analysis using an online target prediction suggested that PELI1 could be a downstream target for miR-142-3p (Fig. 5D-a). Dual-luciferase reporter assays verified that co-transfection of cells with WT-PELI1-3′UTR and miR-142-3p-mimic inhibited luciferase activity compared with that in the corresponding control group (WT-PELI1-3′UTR + NC-mimic) (Fig. 5D-b). No significant changes were observed between the Mut-PELI1-3′UTR + miR-142-3p-mimic and Mut-PELI1-3′UTR + NC-mimic groups. Furthermore, TPC-1 cells were transfected with miR-142-3p mimics to overexpress miR-142-3p (Fig. 5E). The subsequent RT-qPCR analysis showed that miR-142-3p overexpression significantly inhibited PELI1 expression (Fig. 5F). Considering that circRNAs often act as ceRNAs of miRNAs to regulate the expression of their target genes, these results indicated that hsa_circ_0004846 could promote PELI1 expression by inhibiting miR-142-3p expression.

Since the molecular mechanism of hsa_circ_0004846 in the miR-142-3p/PELI1 axis was uncovered, a PELI1-silenced TPC-1 cell line was constructed to further explore the role of this axis in regulating tumorigenesis. The efficiency of cell transfection with PELI1-siRNA was verified by RT-qPCR analysis (Fig. 6A). Furthermore, the CCK-8 assay demonstrated that PELI1 silencing significantly inhibited cell viability in hsa_circ_0004846-overexpressing cells (Fig. 6B). Consistently, PELI1 silencing also reversed the enhanced invasion and migration of hsa_circ_0004846-overexpressing TPC-1 cells (Fig. 6C). In addition, western blotting revealed that the increased expression levels of p-AKT and p-GSK-3β induced by hsa_circ_0004846 were reversed after PELI1 knockdown (Fig. 6D).